
\PassOptionsToPackage{dvipsnames}{xcolor}
\documentclass[12pt, letterpaper]{article}
\usepackage{setspace}
\usepackage[utf8]{inputenc}
\usepackage{pdfpages}
\usepackage{ragged2e} 

\usepackage{graphicx}	% include graphic
\graphicspath{{../graph/},{../../graph/}}	% graphic path
  \usepackage{amsmath}	% math
\usepackage[singlelinecheck=false]{caption}	% table caption
\usepackage{booktabs,caption}				% table note, multicolumn bottom line, \cmidrule{2-3}
\usepackage[flushleft]{threeparttable}		% table note
\usepackage{float}				% table position [H]
\setlength{\tabcolsep}{6pt}     		% General space between columns (6pt standard)
\renewcommand{\arraystretch}{1} 	% General space between rows (1 standard)
\usepackage{rotating}			% landscape/portrait \begin{sidwaysfigure}, \begin{sideways}, \begin{turn}{30}, ...
\usepackage{pdflscape}			% landscape, \begin{landscape} \thispagestyle{empty}
\usepackage{geometry}			% page margin
\usepackage{afterpage}			% page margin for some pages, \afterpage{ \newgeometry{left=10mm, top=10mm} \clearpage \restoregeometry}
\usepackage{makecell}			% column line break, thead{ \\ }
\usepackage{longtable}
\usepackage{subfig}


\usepackage{nopageno}

\usepackage[dvipsnames]{xcolor}



% THIS FILE COMPILES THE TABLES AND FIGURES IN: 

%"Regulation by Reputation? Intermediaries, Labor Abuses, and International Migration" 
% - A. Nilesh Fernando (University of Notre Dame); corresponding author (nilesh.fernando@nd.edu)  
% - Niharika Singh (University of Notre Dame) 
% See readme.docx for details 


\makeatletter

\begin{document}

\pagestyle{plain}

\clearpage


\newpage


% FIGURE 1: EVENT STUDY GRAPHS - ALL AGENCIES 
\newpage
\clearpage
\thispagestyle{empty}
\begin{landscape}

\begin{figure}[htp] 
    \centering
    
    \caption*{\textbf{Figure 1 - Event Study Graphs: All Agencies}}
    \captionsetup[subfigure]{labelformat=empty}
\hspace{25pt}

       
    \subfloat[Panel A: Job Orders]{%
    \includegraphics[scale=0.45]{./job_orders_all_qrt_08.pdf}%
        }%
\hfill
    \subfloat[Panel B: Average Salary]{%
    \includegraphics[scale=0.45]{./salary_all_qrt_08.pdf}%
        }%
 \hfill
    \subfloat[Panel C: Contract Quality]{%
    \includegraphics[scale=0.45]{./contractq_all_qrt_08.pdf}%
   
        }%

\end{figure}
\end{landscape}

\newpage


% FIGURE 2: EVENT STUDY GRAPHS - LOW QUALITY AGENCIES

\newpage
\clearpage
\thispagestyle{empty}
\begin{landscape}

\begin{figure}[htp] 
    \centering
    
    \caption*{\textbf{Figure 2 - Event Study Graphs: Low Quality Agencies}}
    \captionsetup[subfigure]{labelformat=empty}
\hspace{25pt}

       
    \subfloat[Panel A: Job Orders]{%
    \includegraphics[scale=0.45]{./job_orders_low_qrt_08.pdf}%
        }%
\hfill
    \subfloat[Panel B: Average Salary]{%
    \includegraphics[scale=0.45]{./salary_low_qrt_08.pdf}%
        }%
 \hfill
    \subfloat[Panel C: Contract Quality]{%
    \includegraphics[scale=0.45]{./contractq_low_qrt_08.pdf}%
        }%

\end{figure}
\end{landscape}


\newpage

% FIGURE 3: THE DISTRIBUTION OF MATCHES TO GOOD EMPLOYERS BY AGENCY QUALITY



 \begin{landscape}


\begin{figure}[htp] 
    \centering
    
    \caption*{\textbf{Figure 3: The Distribution of Matches to Good Employers by Agency Quality}}
        \label{fig_matches}
        \captionsetup[subfigure]{labelformat=empty}


    \includegraphics[scale=1]{./matches_combo_v3.pdf}%



\begin{justify}
\doublespacing
   \textbf{Notes:} This figure plots probability density functions describing matches between agencies and `Good Employers': employers with a below median pre-program ($<$ 2010) complaint rate. The domain is the continuous Combined Score/2012 agency rating which is only known to the public after 2012. Panel A plots the density function for pre-program matches (dashed gray line) and those during 2010, i.e. the first year of the RI-phase (lighter triangles). Panel B adds 2011, year two of the RI-phase (darker triangles). Panel C plots the remaining years (2012-2015) corresponding to the QR-phase (solid lines, darker lines refer to later years). 
   
\end{justify}
   
\end{figure}
\end{landscape}




\newpage

% TABLE 1 - AGENCY-MONTH LEVEL SUMMARY STATISTICS, 2005-2015 


\begin{table}[H]
\centering

\caption*{Table 1: Agency-Month Level Summary Statistics, 2005-2015}


\input{./table1.tex}


\end{table}


% TABLE 2 - INVESTMENT IN RATING CRITERIA AND MARKET OUTCOMES


\begin{table}[H]
\centering
\caption*{Table 2: Investment in Rating Criteria and Market Outcomes}

\vspace{-0.5em}

\begin{minipage}{\textwidth}
\caption*{Panel A: Effects by Program Phase}
\input{./market_a.tex}
\vspace{1em}

\caption*{Panel B: Effects by Agency Quality}
\input{./market_b.tex}
\vspace{1em}

\caption*{Panel C: Effects by Agency Quality and Program Phase}
\input{./market_c.tex}
\end{minipage}

\end{table}


% TABLE 3 - PLACEMENT QUALITY 

\newpage


\begin{table}[H]
\centering
\caption*{Table 3: Placement Quality}

\vspace{-0.5em}

\begin{minipage}{\textwidth}
\caption*{Panel A: Effects by Program Phase}
\input{./placement_a.tex}
\vspace{1em}

\caption*{Panel B: Effects by Agency Quality}
\input{./placement_b.tex}
\vspace{1em}

\caption*{Panel C: Effects by Agency Quality and Program Phase}
\input{./placement_c.tex}
\end{minipage}

\end{table}



%TABLE 4 - THE EFFECT OF RECEIVING A HIGHER RATING 

\newpage


\begin{table}[H]
\centering
\caption*{Table 4: The Effect of Receiving a Higher Rating}

\begin{landscape}
\input{./table4.tex}
\end{landscape}

\end{table}

%TABLE 5 -  SPILLOVER EFFECTS 

\newpage
\begin{table}[H]
\centering
\caption*{Table 5: Spillover Effects}
\begin{landscape}
\input{./spillovers.tex}
\end{landscape}

\end{table}


%%%%%%%%%%%%%%%%%%%
% ONLINE APPENDIX % 
%%%%%%%%%%%%%%%%%%%




\newpage

%APPENDIX A4 - COMPLAINT TYPES

\begin{landscape}

\begin{table}[H]
\caption*{Appendix A4: Complaint Types}

\scalebox{0.85}{%
\begin{minipage}{\textwidth}
\centering

\input{./complaints_sumstats.tex}
\end{minipage}}
\end{table}
\end{landscape}



% APPENDIX A9: DISTRIBUTION OF RATINGS

\newpage
\begin{landscape}

\clearpage

\begin{figure}[htp] 
    \centering

\subsection*{Appendix A9: Distribution of Ratings} 

    \captionsetup[subfigure]{labelformat=empty}
  
\vspace{30pt}
    \subfloat[Panel A: All Eligible Agencies]{%
    \includegraphics[scale=0.5]{./rating_dist_all.pdf}%
        }%
    \subfloat[Panel B: Recruited 100-155 Migrants in 2009]{%
    \includegraphics[scale=0.5]{./rating_dist_comp_opt.pdf}%
        }%
 
\end{figure}
\end{landscape}



% APPENDIX A10: QUARTERLY VARIATION IN RECRUITMENT 
 
 
\newpage
\begin{figure}[htp] 
    \centering
    
    \caption*{\textbf{\Large{Appendix A10: Quarterly Variation in Recruitment }}}
\vspace{30pt}
    \includegraphics[scale=1]{../out/stochastic_recruitment.pdf}
        \label{fig:a}%

        \vspace{30pt}
    \caption*{\textbf{Note:} This figure shows quarterly recruitment at the agency level for the eligibility period (four quarters in 2009) and two quarters prior to 2009. Attention is restricted here to agencies who recruited between 45 and 155 migrants in 2009, where eligible agencies recruited at least 100 migrants.  }
\end{figure}



% APPENDIX A15: PRE-PROGRAM EMPLOYER TO AGENCY MATCHES

\newpage

\begin{figure}[htp] 
    \centering
    
    \caption*{\textbf{Appendix A15: Pre-Program Employer to Agency Matches}}
    \captionsetup[subfigure]{labelformat=empty}
\hspace{25pt}
    \includegraphics[scale=0.75]{./good_bad_pre.pdf}%
        
\end{figure}



% APPENDIX A16: WAGES AND AMENITIES BY EMPLOYER COMPLAINT RATES


\newpage

\clearpage

\subsection*{Appendix A16: Wages and Amenities by Employer Complaint Rates}


\begin{figure}[htp] 
    \centering
    
\vspace{30pt}


 \subfloat[Panel A: All Employers]{%
    \includegraphics[scale=0.50]{./salaries_amenities_v2_all.pdf}%
        \label{fig:a}%
        }%
    \hfill%
    \subfloat[Panel B: Employers \& Agencies Specializing in Domestic Work]{%
    \includegraphics[scale=0.50]{./salaries_amenities_v2_dw.pdf}%
        \label{fig:b}%
        }%
        
    
          \hfill%

\justify{\textbf{Notes:} This figure displays average salaries (2015 USD) and the share of amenities by pre-program employer complaint rates. The histogram shows the fraction of employers that fall into each complaint rate bin. Panel A restricts attention to employers who recruited at least 50 migrants prior to the program, which accounts for 94\% of all pre-program recruitment. Panel B additionally focuses restricts attention to employers (and/or foreign agencies) for whom at least 75\% of recruited migrants were domestic workers.}

\end{figure}


%APPENDIX A17: MATCHES TO BAD EMPLOYERS BY AGENCY QUALITY


\newpage

\subsection*{Appendix A17: Matches to Bad Employers by Agency Quality}
\label{fig_bad_matches}
  
\begin{figure}[htp!] 
\centering
\includegraphics[scale=0.95]{./bad_matches_v1.pdf}%

\vspace{15pt}

 \justify{\textbf{Notes:} This figure plots a series of probability density functions that describe which recruitment agencies are matched to `Bad Employers': defined as employers who have an above median pre-program ($<$ 2010) complaint rate. A match here is defined as whether an agency sends a migrant to a `Good Employer' in a specific year. The domain is the continuous Combined Score/2012 agency rating which is only known to the public after 2012. Panel A plots the density function for matches occurring prior to the program (\textcolor{red}{red line}) and during the first year of the program (2010), i.e. the \textcolor{blue}{Reputational Incentives (RI)} phase (the \textcolor{blue}{blue line}). Panel B adds 2011, the second year of \textcolor{blue}{RI}. Panel C plots the data for the remaining years (2012-2015) of the \textcolor{OliveGreen}{Quality Revelation (QR)} period (the \textcolor{OliveGreen}{green lines}).}

\end{figure}


\newpage

%APPENDIX A18 - INVESTMENT IN THE AUDIT SCORE

\begin{landscape}

\begin{table}[H]
\caption*{Appendix A18: Investment in the Audit Score}

\scalebox{0.85}{%
\begin{minipage}{\textwidth}
\centering

\input{./audit_invest.tex}
\end{minipage}}
\end{table}
\end{landscape}




\newpage

%APPENDIX A19 - AGENCY EXIT OVER TIME

\begin{table}[H]
\centering
\caption*{Appendix A19 - Agency Exit Over Time }

\input{./firm_exit.tex}


\end{table}



\newpage

%APPENDIX B1 - RESULTS FROM THE AGENCY SURVEY 


\begin{table}[H]
\centering
\caption*{Appendix B1: Results from the Agency Survey}

\vspace{-0.5em}

\begin{minipage}{\textwidth}
\caption*{Panel A: Learning about Employer Reputation}
\input{./agency_survey_a.tex}
\vspace{1em}

\caption*{Panel B: Reasons for Rejecting Job Order}
\input{./agency_survey_b.tex}

\end{minipage}

\end{table}



\newpage

%APPENDIX B2 - CHARACTERISTICS OF JOB ORDERS 


\begin{table}[H]
\caption*{Appendix B2: Characteristics of Job Orders}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: Effects by Program Phase}
\input{./joborder_a.tex}
\vspace{1em}

\caption*{Panel B: Effects by Agency Quality}
\input{./joborder_b.tex}
\vspace{1em}

\caption*{Panel C: Effects by Agency Quality and Program Phase}
\input{./joborder_c.tex}
\end{minipage}}

\end{table}


\newpage

%APPENDIX B3 - COMPONENTS OF CONTRACT QUALITY

\begin{table}[H]
\caption*{Appendix B3: Components of Contract Quality}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: Effects by Program Phase}
\input{./quality_components_a.tex}
\vspace{1em}

\caption*{Panel B: Effects by Agency Quality}
\input{./quality_components_b.tex}

\end{minipage}}
\end{table}



\newpage

%APPENDIX B4 - COMPOSITION OF MIGRANTS RECRUITED

\begin{table}[H]
\caption*{Appendix B4: Composition of Migrants Recruited}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: Effects by Program Phase}
\input{./migrantcomp_a.tex}
\vspace{1em}

\caption*{Panel B: Effects by Agency Quality}
\input{./migrantcomp_b.tex}

\end{minipage}}

\end{table}

\newpage

%APPENDIX B5 - RAW TRENDS

\subsection*{Appendix B5: Raw Trends} 
\begin{figure}[htp!] 
    \centering
    \captionsetup[subfigure]{labelformat=empty}
       
    \subfloat[Panel A: Migrant Recruitment]{%
    \includegraphics[scale=0.50]{./migrant_trend.pdf}%
    }%
\hfill
    \subfloat[Panel B: Job Orders]{%
    \includegraphics[scale=0.50]{./joborder_trend.pdf}%
        }%
 \hfill
    \subfloat[Panel C: Complaint Intensity]{%
    \includegraphics[scale=0.50]{./complaint_trend.pdf}%
        }%

\end{figure}


\newpage

%APPENDIX B6 - COMPONENTS OF QUANTITATIVE SCORE 

\begin{table}[H]
\caption*{Appendix B6: Components of Quantitative Score}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: Effects by Program Phase}
\input{./reputations_a.tex}
\vspace{1em}

\caption*{Panel B: Effects by Agency Quality}
\input{./reputations_b.tex}
\vspace{1em}

\caption*{Panel C: Effects by Agency Quality and Program Phase}
\input{./reputations_c.tex}


\end{minipage}}

\end{table}



% APPENDIX B8: SALARIES PAID BY BAD EMPLOYERS

\newpage

    \subsection*{Appendix B8: Salaries Paid by Bad Employers}


\begin{figure}[htp] 
    \centering
    
\vspace{30pt}

   \includegraphics[scale=0.75]{./bad_employer_salaries_by_elig_status.pdf}
     
        
    
         \hfill%
  \begin{flushleft}
\textbf{Notes:} This figure displays average salaries (2015 USD) paid by `Bad Employers'\textemdash{}those with an above median pre-program ($<$ 2010) complaint rate\textemdash{}to eligible and comparison agencies. The vertical blue line is the point prior at which agencies became aware of the program. The vertical green line shows when ratings were released to the public.

\end{flushleft}
\end{figure}



\newpage

%APPENDIX B9 - HETEROGENEOUS EFFECTS FOR LOW QUALITY AGENCIES 


\begin{table}[H]
\caption*{Appendix B9: Heterogeneous Effects for Low Quality Agencies}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: Experience}
\input{./low_het_age_1.tex}
\vspace{1em}

\caption*{Panel B: Nearby Eligible Agency}
\input{./low_het_density_1.tex}
\vspace{1em}

\caption*{Panel C: Good Employers}
\input{./low_het_good_1.tex}
\end{minipage}}

\end{table}


\newpage

%APPENDIX B10 - SPILLOVERS AND ETHNIC NETWORKS 


\begin{table}[H]
\caption*{Appendix B10: Spillovers and Ethnic Networks}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: 100 ft radius}
\input{./spillover_ethnic_1.tex}
\vspace{1em}

\caption*{Panel B: 500 ft radius}
\input{./spillover_ethnic_2.tex}
\vspace{1em}

\caption*{Panel C: 0.5 km radius}
\input{./spillover_ethnic_3.tex}
\end{minipage}}
\end{table}

\newpage

%APPENDIX B11: K-S TESTS FOR EMPLOYER-AGENCY MATCHES 


\newpage
\begin{table}[H]
\centering
\caption*{Appendix B11: K-S Tests for Employer-Agency Matches}

\input{./ks_output_table.tex}

\end{table}


\newpage

%APPENDIX C1 - PRE-TRENDS

\begin{table}[H]
\centering
\caption*{Appendix C1 - Pre-Trends}

\input{./pretrends.tex}


\end{table}


%APPENDIX C2: PLOTS OF LAGGED EFFECTS 

\newpage
\begin{landscape}

\clearpage

\begin{figure}[htp] 
    \centering

\subsection*{Appendix C2: Plots of Lagged Effects} 

    \captionsetup[subfigure]{labelformat=empty}
  
\vspace{30pt}
    \subfloat[Panel A: Migrants Recruited]{%
    \includegraphics[scale=0.5]{./migrant_lag_qrt.pdf}%
        }%
    \subfloat[Panel B: Avg. Salary]{%
    \includegraphics[scale=0.5]{./salary_lag_qrt.pdf}%
        }%
    \subfloat[Panel C: Contract Renewal]{%
    \includegraphics[scale=0.5]{./renew_lag_qrt.pdf}%
        }%
     \hfill
    \subfloat[Panel D: Job Orders]{%
    \includegraphics[scale=0.5]{./job_orders_lag_qrt.pdf}%
        }%
    \subfloat[Panel E: Good Orders]{%
    \includegraphics[scale=0.5]{./good_orders_lag_qrt.pdf}%
        }%
    \subfloat[Panel F: Contract Quality]{%
    \includegraphics[scale=0.5]{./contractq_lag_qrt.pdf}%
        }%
\end{figure}
\end{landscape}



\newpage

%APPENDIX C3: AGENCY EXIT

\begin{table}[H]
\caption*{Appendix C3: Agency Exit}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: Effects by Program Phase}
\input{./firm_exit_appendix_a.tex}
\vspace{1em}

\caption*{Panel B: Effects by Pre-Program Agency Quality}
\input{./firm_exit_appendix_b.tex}

\end{minipage}}
\end{table}


\newpage

%APPENDIX C4: TOTAL SALARIES

\begin{table}[H]
\caption*{Appendix C4: Total Salaries}

\vspace{-0.5em}

\input{./exit_adjustment_rev.tex}

\end{table}


\newpage


%APPENDIX C5: COMPLAINT RATE 

\begin{table}[H]
\centering
\caption*{Appendix C5: Complaint Rate}

\vspace{-0.5em}

\begin{minipage}{\textwidth}
\caption*{Panel A: Effects by Program Phase}
\input{./complaints_a.tex}
\vspace{1em}

\caption*{Panel B: Effects by Agency Quality}
\input{./complaints_b.tex}
\vspace{1em}

\caption*{Panel C: Effects by Agency Quality and Program Phase}
\input{./complaints_c.tex}
\end{minipage}

\end{table}

\newpage

%APPENDIX C6: ROBUSTNESS TO AGENCY EXIT

\begin{table}[H]
\caption*{Appendix C6: Robustness to Agency Exit}

\vspace{-0.5em}


\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: Effects by Program Phase}
\input{./all_exit_a.tex}
\vspace{1em}

\caption*{Panel B: Effects by Pre-Program Agency Quality}
\input{./all_exit_b.tex}

\end{minipage}
\end{table}

\newpage

%APPENDIX C7: ROBUSTNESS TO TREND CONTROLS 

\begin{table}[H]
\caption*{Appendix C7: Robustness to Trend Controls}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: Quality Trend Only}
\input{./quality_trend_1.tex}
\vspace{1em}

\caption*{Panel B: Recruitment Trend Only}
\input{./pre_recruit_1.tex}
\vspace{1em}

\caption*{Panel C: No Controls}
\input{./no_trends_1.tex}
\end{minipage}}
\end{table}

\newpage

%APPENDIX C8 - MEAN REVERSION AND MEASURES OF AGENCY QUALITY


\begin{table}[H]
\caption*{Appendix C8: Robustness of Estimates to Varying Measures of Pre-Program Agency Quality}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: Predicted Rating}
\input{./combined_het_ps.tex}
\vspace{1em}

\caption*{Panel B: Quantitative Score 2009}
\input{./combined_het_qs_09.tex}
\vspace{1em}

\caption*{Panel C: Avg. Quantitative Score 2007-2009}
\input{./combined_het_qs_0709.tex}
\end{minipage}}
\end{table}

\newpage

%APPENDIX C9 - ROBUSTNESS OF SPILLOVER EFFECTS 

\begin{table}[H]
\caption*{Appendix C9: Robustness of Spillover Effects}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: 100 ft radius}
\input{./spillover_robust_1.tex}
\vspace{1em}

\caption*{Panel B: 500 ft radius}
\input{./spillover_robust_2.tex}
\vspace{1em}

\caption*{Panel C: 0.5 km radius}
\input{./spillover_robust_3.tex}
\end{minipage}}
\end{table}


\newpage

%APPENDIX C11 - PROGRAM EFFECTS WITH ALTERNATIVE SAMPLES

\begin{table}[H]
\caption*{Appendix C11: Program Effects with Alternative Samples}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: Recruited 45-155 Migrants in 2009}
\input{./alt_count_1.tex}
\vspace{1em}

\caption*{Panel B: Recruited 25-175 Migrants in 2009}
\input{./alt_count_2.tex}
\vspace{1em}

\caption*{Panel C: All Agencies}
\input{./alt_count_3.tex}
\end{minipage}}
\end{table}


\newpage

%APPENDIX C12 - ROBUSTNESS OF ESTIMATES TO FBR REFORM

\begin{table}[H]
\caption*{Appendix C12: Robustness of Estimates to FBR Reform}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: Main Specification}
\input{./fbr_robust_1.tex}
\vspace{1em}

\caption*{Panel B: Pre-July, 2013 Data}
\input{./fbr_robust_2.tex}
\vspace{1em}

\caption*{Panel C: Domestic Workers Trend}
\input{./fbr_robust_3.tex}
\end{minipage}}
\end{table}


\newpage

% APPENDIX C13 - RD PLOTS FOR JOB ORDERS

\subsection*{Appendix C13: RD Plots for Job Orders} 
\begin{figure}[htp!] 
    \centering
    \captionsetup[subfigure]{labelformat=empty}
       
    \subfloat[Panel A: First-Order Polynomial]{%
    \includegraphics[scale=0.5]{./c13_a.pdf}%
    }%

    \subfloat[Panel B: Second-Order Polynomial]{%
    \includegraphics[scale=0.5]{./c13_b.pdf}%
    }%
 
    \subfloat[Panel C: Third-Order Polynomial]{%
    \includegraphics[scale=0.5]{./c13_c.pdf}%
	}%
\end{figure}


\newpage

%APPENDIX C14 - ROBUSTNESS OF RD ESTIMATES

\begin{table}[H]
\caption*{Appendix C14: Robustness of RD Estimates}

\vspace{-0.5em}

\scalebox{0.75}{%
\begin{minipage}{\textwidth}
\centering

\caption*{Panel A: First-Order Polynomial}
\input{./table4.tex}
\vspace{1em}

\caption*{Panel B: Second-Order Polynomial}
\input{./c14_a.tex}

\caption*{Panel C: Third-Order Polynomial}
\input{./c14_b.tex}

\end{minipage}}
\end{table}




\end{document} 


